China Mechanical Engineering ›› 2025, Vol. 36 ›› Issue (12): 2986-2992.DOI: 10.3969/j.issn.1004-132X.2025.12.022

Previous Articles    

A Six-axis Robotic Arm Path Planning Based on Improved SAC Algorithm

Jiying TUO(), Xiaonan XU, Jun LI, Yuchen ZHANG, An HUANG, Du HU, Zilin LIU()   

  1. Key Laboratory of Advanced Manufacturing Technology for Automobile Parts(Ministry of Education),Chongqing University of Technology,Chongqing,400054
  • Received:2025-04-09 Online:2025-12-25 Published:2025-12-31
  • Contact: Zilin LIU

一种基于改进SAC算法的六轴机械臂路径规划

妥吉英(), 徐笑南, 李俊, 张玉琛, 黄安, 胡都, 刘梓林()   

  1. 重庆理工大学 汽车零部件先进制造技术教育部重点实验室, 重庆, 400054
  • 通讯作者: 刘梓林
  • 作者简介:妥吉英,男,1988年生,讲师、硕士研究生导师。研究方向为智能控制、非线性动力学。发表论文20余篇。E-mail:tjy@cqut.edu.cn
  • 基金资助:
    重庆市教育委员会科学技术研究计划青年项目(KJQN202201134);重庆市教育委员会科学技术研究计划青年项目(KJQN202101144);重庆市基础科学与前沿技术研究专项(CSTC2020JCYJ-MSXMX0331)

Abstract:

To improve the convergence speed and training stability of the SAC(soft actor-critic) algorithm, an improved SAC algorithm was proposed, incorporating the concepts of advantage functions and reward centering. To validate the performance of the improved SAC algorithm, simulation analyses were conducted in a six-axis robotic arm path planning scenario, comparing with DDPG(deep deterministic policy gradient), TD3(twin delayed deep deterministic policy gradient), and the original SAC algorithm. The results show that the improved SAC outperforms DDPG, TD3, and SAC in both of the convergence speed and stability. After 1500 training episodes, the path planning success rate increases by 4.8% compared to the SAC algorithm. Further experiments confirm the feasibility and effectiveness of the improved SAC algorithm's planning results in real-world environments.

Key words: six-axis robotic arm, path planning, advantage function, reward centering, improved SAC algorithm

摘要:

为提升柔性动作-评价(SAC)算法的收敛速度及训练的稳定性,在引入优势函数与奖励聚中机制的基础上,提出一种改进SAC算法。为验证改进SAC算法的训练效果,在六轴机械臂路径规划场景中进行了仿真分析,并与深度确定性策略梯度(DDPG)算法、双延迟深度确定性策略梯度(TD3)算法及SAC算法进行对比。结果显示,改进SAC算法在收敛速度和稳定性上均超越DDPG、TD3与SAC算法,训练1500回合后其路径规划成功率较SAC算法提高4.8%。进一步的实验验证了改进SAC算法的规划结果在真实环境中的可行性与有效性。

关键词: 六轴机械臂, 路径规划, 优势函数, 奖励聚中, 改进SAC算法

CLC Number: